The federal Open Payments database makes public the amount and type of payments from drug and device companies to individual physicians. It seems plausible that very large payments warrant greater concern about conflict of interest causing bias and undue influence. The degree of concern will vary, however, according to the type of financial relationship and the nature of the physician’s advice or activity. In several situations, the existence of bias can be investigated directly, or the financial relationship effectively and simply managed. Very Large Payments for Royalties, Licenses, or Sale of Companies The 5 physicians who received the largest total payments in the Open Payments database from 2013 through 2015 each received more than $28 million. These physicians were inventors, owners, and officials of companies that were purchased by the company reporting the payment. What are the implications of identifying these financial relationships? Such large payments raise significant concerns about the likelihood of bias or undue influence in committees, such as those that determine operating room supplies, hospital formularies, or clinical practice guidelines, whose decisions or recommendations affect many patients. It seems prudent to exclude physicians with a large financial stake in a drug or device from these committees. Excluding them would not impair the committees’ primary interest in improving the quality of patient care. To gain the expertise of such physicians, committees can invite them to present their views, without participating in deliberations or voting. In contrast, physicians whose only payments from industry are small amounts for meals would seem to raise fewer a priori concerns about undue influence or bias in such committees. Large payments for royalties and licenses, as well as stock options, also raise concerns about undue influence and bias in clinical trials. The value of these financial relationships in the company sponsoring the trial would increase, perhaps substantially, if clinical trial results are positive. Concerns arise if persons with these relationships serve as principal investigator or determine participant outcomes because these roles involve great discretion and many implicit judgments in which conscious or unconscious bias could enter. Concerns are reduced if experts without such relationships fill these key roles, whereas the physicians receiving such large payments could still provide expertise as a consultant. It seems very unlikely that such a policy would impair the validity of a clinical trial, which is the primary interest in clinical research. Very Large Payments for Promotional Talks The 5 physicians who received the greatest number of payments from August 2013 to December 2015 in the Open Payments database received a median of 207 payments from 27 companies for “speaking, training, and education engagements that are not part of continuing education.” Their median total payment was $570 000. These engagements included dinner talks at restaurants and satellite presentations at professional society meetings. Such promotional activities are not subject to requirements from the Accreditation Council for Continuing Medical Education to give a “balanced view of therapeutic options” and to “promote improvements or quality in healthcare and not a specific proprietary business interest of a commercial interest.” Promotional talks aim to increase product sales and may bias the prescribing practices of clinicians. It seems unlikely that these speakers have the time and intellectual bandwidth to prepare and revise so many different talks for multiple products and companies. It is much more likely that the product sponsor or a communications consultant hired by the sponsor prepares the promotional talks and slides. Physicians who give a very large number of promotional talks may also speak at accredited continuing medical education (CME) programs, which are required for renewal of licenses or to maintain staff privileges. Common-sense skepticism should stimulate particularly careful review of presentations by speakers who also give a very large number of promotional talks. Questions to ask include: Does the talk inappropriately recommend enlarging the population to be treated by a drug? Overstate the clinical significance of the product’s benefits? Underemphasize adverse effects? Downplay alternative approaches and comparative costs? Slides from other speakers, including those who received much smaller payments for meals, also are routinely reviewed in accredited CME, but seem to raise fewer a priori concerns about bias than physicians who receive very large payments. Perhaps individuals who approve talks that are accredited CME should be aware of the payments made to those speakers at the time of the review. How Large Is Large? How much is very large, and at what point do large payments clearly imply an unacceptable likelihood of bias or undue influence? $10 000 per year? $50 000? $100 000? Currently no data exist to answer this question. To help guide conflict of interest policies, researchers should use the Open Payments database, together with other available data, such as speakers’ slide decks, to evaluate the relationship between the size of payments and bias in presentations. The type of payment matters as well as the amount. Payments for inventions leading to treatments, for developing effective new therapies, and for valid research serve the primary interest of patients for effective and safe therapies and as such should be encouraged. At the other extreme, some types of payments seriously undermine the primary interests of patients or medical institutions. For example, the primary goal of medical schools and hospitals with residency and fellowship programs is to train learners to critically evaluate clinical and scientific evidence, an essential skill for lifelong learning and professionalism. Allowing faculty to be paid for promotional talks prepared primarily by companies contradicts the expectations for students and trainees to think critically and receive credit only for work to which they made significant contributions. Several medical schools do not permit faculty to join speakers’ bureaus. This ban should be universal. Small Payments Some very small gifts also matter enough to justify restrictions that have few adverse consequences. According to a recent study, when the Open Payments database was linked to databases on physician prescriptions to Medicare patients, there was a significant association between physicians receiving a meal promoting 1 of 4 brand-name drugs and higher rates of prescribing the promoted drug compared with alternative generic drugs in the same class. The promoted drugs were brand-name statins, angiotensin receptor blockers, cardioselective β-blockers, and selective serotonin or norepinephrine reuptake inhibitors, for which there is only limited, mixed, or contrary evidence that they are superior to generic equivalents. The vast majority of physicians received a single meal, costing generally between $12 and $18. The study also found that receiving more meals or meals costing more than $20 was associated with even higher rates of prescribing. Because this study used data on payments for an entire year, but data on meals for only the last 5 months of the year, it is possible that physicians who accept meals already favored the marketed drugs, rather than that a single promotional meal influences prescribing. However, even if this explanation accounts for the association, it is undesirable to reinforce prescribing patterns that lack an evidence-based rationale yet increase costs, including out-of-pocket costs to patients. Defenders of promotional meals argue that there is no definitive evidence that that they harm patients. However, this absence of evidence does not establish a lack of harm. Moreover, no empirical studies show that meals from drug manufacturers improve patient outcomes. On the basis of experimental studies with nonphysicians, the social science and marketing literature concludes that even small gifts induce unconscious feelings of gratitude and reciprocity. Drug company representatives are trained to use small gifts, such as meals, to develop relationships with physicians and their staff and to persuade physicians to prescribe a target drug. A small gift to an individual physician may influence prescribing by only that physician, but such gifts are ubiquitous. In one study, more than half of US physicians received payments for meals. This evidence led many medical institutions to ban onsite company-sponsored meals and branded promotional items (such as pens or notepads), and led pharmaceutical companies to stop offering such branded items. Conclusions Very high payments for certain purposes raise concerns about conflicts of interest in situations that can and should be prohibited or effectively managed. Questions about smaller payments cannot be answered definitely with current information. Additional research using the Open Payments database with innovative designs could provide policy-relevant information. Ultimately, however, the management of financial conflicts of interest requires judgments that should take into account not only the size of payments, but also the type of payment, the type of physician activity or decision at stake and their benefits, how bias or undue influence is likely to affect the medical profession’s primary interest in the activity, and the effectiveness and feasibility of managing the conflicts of interest. Source
